Camelia, the Perl 6 bug

IRC log for #mojo, 2010-09-28

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:35 ashleyde1 left #mojo
00:35 ashleydev joined #mojo
00:39 stereonaut joined #mojo
01:05 stereonaut left #mojo
01:08 perlrocks Twitter: "Mojolicious, a framework for Perl - http://is.gd/fwBOL" (es) --fr33co http://twitter.com/fr33co/status/25740533041
01:11 perlrocks Twitter: "jajajajaja good name RT @ fr33co Mojolicious, a framework for Perl - http://is.gd/fwBOL" (es) --negrunch http://twitter.com/negrunch/status/25740798764
01:15 polvo left #mojo
02:05 GabrielVieira joined #mojo
02:27 GabrielVieira left #mojo
02:27 GabrielVieira2 joined #mojo
02:35 janus left #mojo
02:48 janus joined #mojo
02:50 stereonaut joined #mojo
03:42 Foxcool joined #mojo
04:03 tl joined #mojo
04:05 tholen42 left #mojo
04:45 ash_ left #mojo
05:19 stereonaut left #mojo
05:24 ysyrota joined #mojo
05:43 koban joined #mojo
05:47 Foxcool left #mojo
05:48 Foxcool joined #mojo
05:54 kvorg left #mojo
06:09 yakudza joined #mojo
06:17 kvorg joined #mojo
06:33 esskar left #mojo
06:43 esskar joined #mojo
06:49 esskar moin
06:50 yakudza left #mojo
06:58 esskar__ joined #mojo
07:01 esskar left #mojo
07:02 esskar__ is now known as esskar
07:02 doubi joined #mojo
07:26 marcus_ moin
07:34 ltriant left #mojo
07:37 doubi esskar_: ping?
07:39 spleenjack joined #mojo
08:14 MojoGuest979 joined #mojo
08:14 MojoGuest979 From: http://www.google.com.tw/search?hl=zh-TW&sourc​e=hp&biw=1680&bih=989&q=websocket+demo​&aq=f&aqi=g1&aql=&oq=&gs_rfai= (1 hits)
08:16 su-bzero_ joined #mojo
08:21 su-bzero left #mojo
08:24 koban left #mojo
08:24 koban joined #mojo
08:37 MojoGuest979 xcs
08:37 MojoGuest979 dddd
08:37 MojoGuest979 eee
08:37 MojoGuest979 left #mojo
08:39 doubi Morning all. Anyone know if Luzy supports multi-language partials?
08:48 doubi That is, providing an easier way than rolling my own of having different sections of a page in different languages.
08:53 kvorg left #mojo
09:08 Rhaen moin alls!
09:09 Rhaen fff
09:10 Rhaen gggg
09:13 vti Rhaen: what do you smoke?
09:14 Rhaen vti: why do ask? If you need something, ask me in a query about it
09:25 marcus_ :D
09:25 sri moin
09:25 sri lol
09:26 marcus_ mojo - because drugs alone are not enough.
09:33 MojoGuest732 joined #mojo
09:33 MojoGuest732 From: http://www.google.com/search?sourceid=c​hrome&ie=UTF-8&q=websocket+demo (3 hits)
09:34 MojoGuest732 left #mojo
09:46 MojoGuest154 joined #mojo
09:46 MojoGuest154 From: http://www.google.co.uk/search?sourceid=c​hrome&ie=UTF-8&q=websockets+demo (1 hits)
09:46 MojoGuest154 left #mojo
09:49 jest joined #mojo
09:56 esskar doubi: explain
09:57 esskar you mean different languages in one file?
09:57 esskar i thought abou that
09:57 esskar i thought about that
09:58 esskar but that would make it harder for database stores
09:58 doubi esskar: Sure it would. That's why I hoped you'd done it for me :)
09:58 esskar whats wrong with having a file for every language?
09:59 esskar i'm working on roles and users for luzy (thats one reason for MojoX::Authen)
10:00 esskar then, i will make the admin interface simpler
10:00 doubi I have pages giving information about community groups. Users will want to provide alternate language entries for certain parts of that information, e.g. the group description, but not others, e.g. fields about dates and times of events, which can be translated automatically.
10:00 esskar so, you could use that to manage your content
10:01 esskar thats more an i18n thing, isnt it?
10:02 doubi Yea, maybe I'm not using the right terms in asking my question...
10:02 doubi A complete page for each language would duplicate quite a bit of information
10:03 doubi What I want is to check for a certain translation for individual pieces of content on one page, falling back to English or whatever if the community group hasn't provided a translation in the requested language
10:03 esskar i will think abou that
10:03 esskar but i think you can use the resolver for that
10:04 doubi How d'you mean?
10:04 esskar http://github.com/esskar/luzy/blob/maste​r/lib/Mojolicious/Plugin/Cms/Resolver.pm
10:05 MojoGuest319 joined #mojo
10:05 esskar you could define a resolver, for example
10:06 esskar i18n
10:06 esskar and you write
10:06 esskar <i18n pattern="foo" />
10:06 esskar <i18n pattern="bar" />
10:06 esskar <i18n pattern="uhu" />
10:06 esskar in your content
10:06 esskar then you bin i18n to an something that reada the attribute pattern
10:07 esskar check if you have a translation for that think in pattern
10:07 esskar or take the default
10:07 esskar see the "Twitter search" example
10:08 doubi Sounds cool. Thanks for the tutelage :)
10:12 esskar doubi: one hint, the resolver expects valid XHTML
10:13 doubi Thanks. I do try to close my single-tag elements and such :-)
10:15 sri if he's using Mojo::DOM that doesn't matter
10:19 doubi sri: I've seen some of the fun one-liners you can do with Mojo::DOM & -Mojo, and it looks like a cool alternative to Web::Scraper. Is that the kind of thing it's used for in Mojolicious apps too, pulling in content, or does it have an ulterior motive?
10:20 sri i just wanted to build the most pleasant to use xml parser ever
10:20 sri because usually dealing with xml is painful
10:20 sri not anymore :)
10:21 doubi A worthy cause indeed :)
10:21 sri jquery showed us that the answer is css selectors
10:26 vti sri showed us the way
10:26 doubi W::S uses a similar interface, with many more dependencies of course.
10:27 sri scraping is just one of the uses, another goal was versatility
10:27 sri the usual mojo building block
10:27 doubi Ah, I guess so.
10:27 sri like tempire has used it to implement google voice i think
10:28 doubi Mojo::DOM? What does it do that enables a VoIP service?
10:28 sri it parses xml
10:28 sri :)
10:28 a|newkirk[assoc] see Scrappy
10:32 GitHub153 joined #mojo
10:32 GitHub153 mojo: master Sebastian Riedel * 74763dd (20 files in 10 dirs): deprecated finished and receive_message in favor of on_finish and on_message - http://bit.ly/9A3c6d
10:32 GitHub153 left #mojo
10:32 MojoGuest616 joined #mojo
10:32 MojoGuest616 From: http://www.google.no/search?aq=f&amp;sou​rceid=chrome&amp;client=ubuntu&amp;chann​el=cs&amp;ie=UTF-8&amp;q=websockets+demo (1 hits)
10:32 MojoGuest616 left #mojo
10:35 sri on_message looks quite a bit better imo
10:36 vti indeed
10:37 sri a|newkirk[assoc]: init and var seem a little out of place, otherwise i like the DSL
10:39 a|newkirk[assoc] thanks, sri, i wrote an article going over the feature set, it looks modest at first glance but it does resume, session handling, etc
10:39 MojoGuest319 left #mojo
10:44 a|newkirk[assoc] sri, i didnt get to my work this weekend but im settling down to work on the local lib thing today...
10:45 a|newkirk[assoc] hows about Mojo::Command::Generate::LocalLib?
10:45 sri maybe, arguable if it's a generator command or not
10:46 sri Mojolicious::Command::LocalLib might be better
10:46 a|newkirk[assoc] oh i see
10:46 sri (or Mojo::Command::LocalLib)
10:47 a|newkirk[assoc] are those namespaces aggregated?
10:47 sri yes
10:47 a|newkirk[assoc] okay
10:47 sri all generator commands have been moved to mojolicious though
10:47 sri since some people got confused
10:49 a|newkirk[assoc] is "mojo package" too ambiguous
10:50 sri or maybe it is a generator command :)
10:50 a|newkirk[assoc] mojolicious boxit :}
10:52 a|newkirk[assoc] how cool would that be ... mojolicious generate app WebApp boxit
10:52 sri rails people call it freezing :)
10:53 a|newkirk[assoc] they do? mmm
10:53 sri rake rails:freeze or so
10:53 sri then it all goes into the vendor directory
10:54 a|newkirk[assoc] hahahahah
10:54 a|newkirk[assoc] i did mojolicious generate app --help
10:54 a|newkirk[assoc] and it create an app under --help dir
10:54 sri lol
10:54 a|newkirk[assoc] then i did rm -Rf --help
10:54 a|newkirk[assoc] and it displayed the help
10:54 a|newkirk[assoc] :)
10:54 sri xD
10:55 sri wonder if we can handle the --help case better
10:57 a|newkirk[assoc] i have no idea how to delete this dir >:|
10:57 sri backslash escape
10:58 sri aftr \- it should even autocomplete
10:59 a|newkirk[assoc] im on windows so i cheated with rmdir /S
10:59 a|newkirk[assoc] sri, have you ever looked at App-Rad?
11:00 sri nope
11:00 a|newkirk[assoc] --help is convention though, maybe just for the app generator we should account for that goof
11:13 sri strange i've not seen App::Rad, it is so similar
11:17 sri ah, it was written about the same time as Mojo::Commmands
11:18 sri and the command list i bet is also inspired by git :)
11:26 a|newkirk[assoc] yup, i help on that project with garum, he's an avid Mojo user
11:27 sri i know, strange he never mentioned it
11:30 DaTa a|newkirk[assoc]: rm -rf ./--help
11:30 a|newkirk[assoc] thanx DaTa, i got it
11:31 a|newkirk[assoc] sri, the project has been inactive for about 6-8 months, i plan to get back to it on my off-season
11:32 a|newkirk[assoc] also, i'll probably take concepts from Mojo::Command as App::Rad does currently support multiple controllers
11:43 doubi The Makefile.PL for Luzy creates the Cms plugin, not one called 'Luzy'... should I be using "plugin Cms =>" instead? I think I'm missing something rather basic...
11:44 crab what is luzy?
11:45 a|newkirk[assoc] Mojo::Command, no prompt, no read_file, huh?
11:45 a|newkirk[assoc] i see write_file
12:03 sri a|newkirk[assoc]: right, no prompt yet, trying to have good defaults that can be changed with command line arguments
12:03 sri most users hate prompts
12:04 sri cpanp vs cpanm ;)
12:26 a|newkirk[assoc] ah
12:32 tempire left #mojo
12:39 doubi crab: http://github.com/esskar/luzy
12:49 Foxcool left #mojo
12:50 yko joined #mojo
12:54 esskar doubi: Luzy is a wrapper around Cms
12:54 esskar i recommend using Luzy
12:54 esskar -d +t
12:56 esskar a|newkirk[assoc]: your nick is too long ;)
13:18 * sri agrees with esskar, and it doesn't even work well with auto complete
13:18 sri all the cool kids have 3 letter nicks ;)
13:19 esskar ;)
13:20 sri oh, yko is back :o
13:22 a|newkirk[assoc] okay, i'll fix that
13:22 esskar sri: can make Test::Mojo "send" cookies on subsequent gets, puts
13:22 sri it does
13:22 sri out of the box
13:22 esskar hmm
13:22 esskar wait
13:23 sri has good tests
13:23 a|newkirk[assoc] is now known as ana
13:23 esskar ah
13:23 esskar it looks at the path ;)
13:27 DaTa ana: thanks :)
13:29 ana np
13:35 ana cant believe sri got me with the ole "all the cool kids are doing x"
13:36 sri sneaky sri
13:36 marcus_ ana: all the cool kids are doing ecstacy?
13:36 ana ha
13:36 ana do we have a pastebot?
13:36 ana and service?
13:37 sri multiple services but no bot i think
13:37 esskar http://github.com/esskar/mojox-authen/commi​t/6f1f35d6bc7a8b5a2bde7853004f5029c9d29eb9
13:38 arthas joined #mojo
13:39 marcus_ <3 gist
13:39 ana i'll just use http://paste.scsys.co.uk/ if thats cool
13:39 marcus_ it's not cool, but it works.
13:39 marcus_ ;)
13:40 ana lol
13:40 * sri prefers gist too
13:41 marcus_ gist + textmate integration = <3
13:42 ana Mojolicious::Command::Package - http://gist.github.com/601008
13:42 ana that was easy
13:42 ana sorta
13:42 perlrocks Twitter: "Form authentication with MojoX::Authen just works #mojolicious http://github.com/esskar/mojox-authen" --esskar http://twitter.com/esskar/status/25785376474
13:43 sri esskar: that looks easy :)
13:44 sri esskar: does it produce good log messages?
13:44 esskar nope
13:46 sri ana: neat
13:47 ana since they point of Mojolicious::Command::Package is to make the app self-contained, any objections to include a generic .htaccess file?
13:47 ana sri, thanks
13:48 doubi esskar: I understand that Luzy is a wrapper around Cms - I'm just being dull in setting it up <:-)
13:48 sri think i would split them in two commands
13:48 esskar doubi: whats the prob?
13:48 doubi git pull... make install installs Cms... Lite app complains plugin Luzy not found... ?
13:48 esskar it does not install the Luzy stuff?
13:49 doubi No, 'cause that's just in /lib/Luzy, not /lib/Mojolicious/Plugin/Luzy
13:49 esskar let me see
13:50 doubi So I should just move it there and install again?
13:50 esskar nope
13:51 esskar doubi: do you have the newest mojo from github?
13:52 doubi esskar: Ahh, only from CPAN, not github. Hang on and I'll update...
13:52 esskar yep
13:52 esskar thats requirement
13:53 doubi Missed that, sorry
13:54 esskar no worries
13:55 und3f[A] is now known as und3f
13:55 perlrocks Twitter: "RT @esskar: Form authentication with MojoX::Authen just works #mojolicious http://github.com/esskar/mojox-authen" --kraih http://twitter.com/kraih/status/25786459567
14:01 crab form authentication?
14:01 sri basic, digest...
14:01 esskar and form
14:02 crab what does form authentication mean?
14:02 esskar you have an html form
14:02 esskar with username + password fields
14:02 crab like <form><input name=login><input name=password><input type=submit></form>?
14:02 crab ah
14:02 esskar yau
14:06 ana shyte, forgot to grab Mojo.pm and Mojolicious.pm
14:06 sri MojoX?
14:10 perlrocks Twitter: "alnewkirk created gist: 601008:
14:11 sri Oo
14:11 ana double-shyte
14:11 * sri kicks perlrocks
14:11 ana thats an easy fix though
14:11 sri you broke perlrocks somehow
14:12 ana Mojo.pm Mojolicious.pm and the MojoX dir I didnt account for
14:12 ana easy fix, brb
14:13 stereonaut joined #mojo
14:24 ana fixed http://gist.github.com/601074
14:25 ana sri, what about generating an .htaccess file?
14:25 sri separate generator command?
14:26 ana ? ahdunno
14:26 ana ... and we're self-contained ladies and gents http://awnstudio.com/blah/script/blah.cgi/
14:26 sri btw. there is an easier solution for the whole bundling thing
14:27 ana really?
14:27 ana whaty
14:27 ana what*
14:27 sri the latest mojolicious version is always available at http://mojolicio.us/Mojolicious-latest.tar.gz and some github download url
14:27 ana ha, nice
14:28 ana but thats the library, not a scaffold right?
14:28 sri it's the normal distribution right
14:28 sri but you could bundle the real thing
14:29 sri like add a mojo dir and have the generated scaffold use lib 'mojo/lib'; or so
14:29 sri little more like rails freezing
14:30 ana whats the difference between doing that and the command I just wrote except one is executed automatically form the scaffold script?
14:31 sri yours might include unrelated files in the namespace
14:31 sri it does not include scripts
14:31 vti git submodules is good enough
14:31 sri thats true of course if you develop with git
14:32 sri vti: didn't you blog about git submodule?
14:32 sri right, git submodule makes even more sense, since this all happens on developer machines
14:33 vti no
14:33 ana wait, whats happening :}
14:33 ana uh
14:33 vti and updates are easy
14:33 ana i took care to only use Modules that are core Perl 5.8
14:34 sri http://github.com/guides/using-​git-submodules-to-track-plugins
14:34 sri you can even get fresh git updates that way
14:36 koban left #mojo
14:36 ana isnt that in a sense, adding a dependency
14:36 sri not really, you fetch it on the developer machine
14:37 sri and deploy the whole thing at once
14:37 ana also, all the cool kids use git
14:37 sri :D
14:37 ana this bypasses CPAN kinda sorta
14:38 sri aye
14:38 ana actually it does, not kinda sorta
14:38 sri brave new open source world
14:38 ana sri, your speaking my language
14:39 esskar http://www.heise.de/newsticker/meldung/​Metro-verkauft-iPhone-3GS-1097712.html *g*
14:39 ana so abandon my Perlish traditions, through away my cpan shackles and run, far and fast
14:39 vti http://babyl.dyndns.org/techblog/entry/galuga
14:39 vti The answer to all your Perl blogin needs. :) http://getbootylicious.org!!!
14:40 ana i like this page http://getbootylicious.org​/pages/documentation.html
14:40 sri cpan should be based on git
14:40 ana hahaha
14:41 ana im too am a rebel-rouser sri
14:41 ana ;)
14:41 perlrocks Twitter: "alnewkirk created gist: 601074:
14:41 ana webapps shouldnt be installed like traditional CPAN modules IMO
14:41 ana this is why and where other languages have a leg up
14:41 vti hm.. parser is broken :(
14:42 sri ana: how did you break perlrocks?
14:42 ana cakephp (shit) but, wget, untar, chmod, start coding
14:42 ana ^^libs includes
14:42 ana included
14:42 ana i did?
14:43 sri see broken tweet above
14:43 ana ah
14:43 ana no idea, that thing hates me
14:44 sri hmm, does that tween include newlines?
14:44 sri *tweet
14:44 ana nope, heres the gist, http://gist.github.com/601074
14:45 sri yes it does
14:45 sri http://twitter.com/alnewki​rkassoc/status/25790565384
14:45 ana where?
14:45 sri shows up in Tweetie
14:45 ana oh, thats github tweeting for me? i guess, wtf?
14:45 sri lol
14:47 ana check out my new website (sneak peek/not finished), let me know what y'all think when yuse have time , http://ana.im
14:48 perlrocks left #mojo
14:48 esskar whats .im ?
14:48 perlrocks joined #mojo
14:48 ana .im == internet mastermind
14:48 ana or something similar
14:48 DaTa dig txt im.tld.pilgerer.org
14:49 sri damn, 3 letter domain, now i'm jealous
14:50 esskar http://sri.im/ ;)
14:50 ana sri.it
14:50 vti sri.im sounds funny in russian
14:50 tempire_ joined #mojo
14:51 ana i tried to get an .it or .io but no such luck
14:51 * tempire_ kicks esskar
14:51 tempire_ readme!
14:51 sri hmm, godaddy wants to sell me thesri.com -.-
14:52 ana :?
14:52 vti sri-is-aweso.me
14:52 esskar tempire_: not yet ;)
14:52 tempire_ you would have to get a more menacing avatar if you use that domain.
14:52 sri like the HAL eye?
14:52 ana cri.me
14:53 ana actually i came really close to registering she.ban.gs for the fuck of it
14:53 ana and it wasnt that much
14:54 esskar sri.ae is stil available
14:54 ana .ng domains start at 4K, pass
14:55 esskar .ae is ok
14:55 esskar but you have to go to their nic office
14:55 esskar no internet subscriptioon available
14:55 sri xD
14:59 ana cant wait to get my ticket to perl oasis
14:59 esskar perl oasis?
15:00 esskar orlando?
15:00 doubi left #mojo
15:01 sri oh stupid.jobs is still available
15:01 ana esskar, yup
15:02 sri ah $90 per year for a .jobs domain
15:03 crab is steve.jobs available?
15:03 ana nice
15:04 ana http://self.jobs
15:05 * ana goes to watch Dexter OnDemand with the wife
15:05 esskar ana: have fun
15:05 sri oh right dexter is back
15:05 esskar whatched simpsons last night
15:05 sri yea, have fun, and don't spoil it later :)
15:06 esskar and "bored to death"
15:06 sri no steve.jobs
15:06 sri but steve.asia Oo
15:06 sri lol steves.jobs is available
15:06 esskar STEVES.JOBS is avail ;)
15:06 sri just saw it :D
15:07 sri freesteve.jobs
15:07 sri also a neat suggestion
15:07 esskar :D
15:07 tempire_ fakesteve.jobs also
15:09 tempire_ someone should register perl.jobs, put up a mojolicious site, and post how awesome mojolicious is all over the job site.
15:09 esskar tempire_: you do it!
15:10 tempire_ I'll do it if you promise to make readme's first in all your projects, and encourage everyone else to do the same.
15:10 esskar i'm not a poet
15:12 esskar i just hated writing essays at school
15:13 tempire_ All you have to do is write a blurb of what it does, and include the pod of your main module.
15:13 tempire_ doesn't have to be pretty
15:13 tempire_ that's my job ;)
15:14 crab i can't clone mojox-authen
15:14 crab stupid github keeps resetting the connection
15:23 esskar__ joined #mojo
15:26 tempire_ left #mojo
15:27 esskar left #mojo
15:27 esskar__ is now known as esskar
15:31 ysyrota left #mojo
15:50 Foxcool joined #mojo
16:06 koban joined #mojo
16:06 jest left #mojo
16:10 doubi joined #mojo
16:22 sri pizzatime!
16:26 yko left #mojo
16:28 kimoto left #mojo
16:32 spleenjack left #mojo
16:35 crab is that like hammer time?
16:38 arthas left #mojo
16:42 ^stereonaut joined #mojo
16:42 doubi crab, Stop.
16:46 sri just like hammer time, but more tasty!
16:47 stereonaut left #mojo
16:47 vti sri: https://gist.github.com/36e4f7c102ae174b3d3f
16:48 vti old bug =/
16:48 vti i will push if you don't mind :)
16:49 sri sure
16:50 GitHub130 joined #mojo
16:50 GitHub130 mojo: master vti * 2ee9529 (3 files in 3 dirs): Fixed pod renderer bug - http://bit.ly/cJ5uYY
16:50 GitHub130 left #mojo
16:51 sri oh noes
16:51 vti what?
16:51 sri the (vti) needs to be after the "."!
16:51 vti f*ck
16:51 sri !@#$%
16:52 GitHub92 joined #mojo
16:52 GitHub92 mojo: master vti * aed88aa (1 files in 1 dirs): Fixed Changes - http://bit.ly/bLKxYN
16:52 GitHub92 left #mojo
16:52 * sri will pretend that never happened
16:52 * vti bows
16:54 vti now it works! http://getbootylicious.org​/pages/documentation.html
16:54 sri neat
17:00 Drossel joined #mojo
17:01 GitHub9 joined #mojo
17:01 GitHub9 mojo: master Sebastian Riedel * 8ad7f8b (1 files in 1 dirs): test cleanup - http://bit.ly/9LFlNV
17:01 GitHub9 left #mojo
17:01 sri regex was a bit too long
17:01 Kulag left #mojo
17:02 sri really wonder why nobody built a cool pod browser yet with mojolicious
17:03 sri maybe too easy :o
17:11 ash_ joined #mojo
17:13 yakudza joined #mojo
17:21 Foxcool left #mojo
17:26 yakudza left #mojo
17:35 esskar left #mojo
17:35 esskar joined #mojo
17:35 koban left #mojo
17:37 GitHub54 joined #mojo
17:37 GitHub54 mojo: master Sebastian Riedel * 8bd773e (18 files in 9 dirs): deprecated process method in Mojo::Client and added new start method - http://bit.ly/dgQ2La
17:37 GitHub54 left #mojo
17:37 sri more consistent
17:50 spleenjack joined #mojo
17:51 crab esskar: your module does authentication not authorization (for the docs)
17:59 ash_ left #mojo
18:03 koban joined #mojo
18:11 doubi left #mojo
18:22 tempire_ joined #mojo
18:22 tempire_ podilicious
18:38 su-bzero joined #mojo
18:40 esskar_ crab: yep, thanks
18:40 sri btw. is anyone having problems with the current client api?
18:41 sri maybe stuff that doesn't make sense?
18:42 su-bzero_ left #mojo
18:42 sri everybody should have trouble with the whole sync/async thing, because it's confusing by nature :)
18:51 und3f sri, it is not so easy to write sync tcp clients on Mojo::IOLoop
18:51 und3f but async is ok
18:52 sri how come?
18:52 sri sync should be trivial
18:52 sri Mojo::IOLoop->new->connect(...) # look ma, blocking!
18:54 sri well, you have to take care of start/stop, but thats in the nature of the reactor pattern
19:00 esskar_ ana, btw, i know who you are! ;-)
19:00 ash_ joined #mojo
19:01 sri ELMO KNOWS WHERE YOU LIVE!
19:01 esskar_ sri: you should add $client->process( $req )
19:01 sri esskar_: no i shouldn't
19:02 esskar_ why not?
19:02 sri why should i?
19:04 esskar_ it would make my pause example work (i think)
19:04 Htbaa joined #mojo
19:04 sri http://www.youtube.com/watch?v=qSoFRIHtUbM
19:06 sri esskar_: ?
19:06 sri you might be misunderstanding something
19:07 esskar_ sure!
19:28 ash_ left #mojo
19:31 ash_ joined #mojo
19:42 Rhaen_ joined #mojo
19:42 Rhaen left #mojo
19:42 Rhaen_ is now known as Rhaen
20:10 tempire_ sri: stuff that doesn't make sense?  as in, trying to post to a url and it just fails for no particular reason?
20:10 sri tempire_: no
20:11 tempire_ then no.
20:11 sri and otherwise?
20:15 tempire_ Mojo::Client ssl transactions fail silently with IO::Socket::SSL
20:16 tempire_ unrelated, but _very_ annoying.
20:16 sri example?
20:16 sri odyssey:mojo sri$ ./script/mojo get https://amazon.com:80
20:16 sri Couldn't open page "https://amazon.com:80". (SSL connect attempt failed because of handshake problemserror:00000000:lib(0):func(0):reason(0))
20:16 sri thats not silent at all
20:17 tempire_ sorry, didn't include the most relevant part.
20:17 tempire_ with IO::Socket::SSL 1.30 (current is 1.33)
20:17 tempire_ it's not related to mojolicious, I know.
20:17 sri odyssey:mojo sri$ MOJO_NO_TLS=1 ./script/mojo get https://amazon.com:80
20:17 sri Couldn't open page "https://amazon.com:80". (Couldn't connect.)
20:18 sri not silent
20:19 tempire_ only with IO :: Socket :: SSL 1.30
20:20 sri quite sure that is equal to the env var
20:21 tempire_ I've already updated it, and previous versions aren't available on cpan.org.  I'll provide an example later.
20:22 tempire_ ah, there we go.
20:41 esskar_ sri, why not supporting  $client->process( $req )
20:41 sri esskar_: why support it?
20:41 esskar_ or add a  $client->request( $req )
20:41 sri give me a reason
20:42 esskar_ so, you could re-request an request (just adding a little header=
20:42 esskar_ -n -= +)
20:42 sri huh?
20:42 sri i don't get it
20:42 sri how would that api change anything?
20:43 esskar_ remember my example from sunday
20:43 sri no
20:43 esskar_ the one with basic auth and pause.perl
20:43 sri and i'm quite sure i told you that rerunning requests reliably is almost impossible
20:44 esskar_ i thought it had someting to do with the transaction
20:44 sri because we allow dynamic content
20:44 ash_ left #mojo
20:44 sri no, it has nothing to do with the transaction
20:44 sri it is the message
20:45 sri you can't reliably build the same http message twice
20:45 esskar_ but a request is only dynamic until its first build
20:45 esskar_ after that, it is actually a static thing
20:45 esskar_ you even return it in the callback
20:46 esskar_ its already build and send
20:46 esskar_ resending the same thin (with different headers) can make sense
20:46 esskar_ +g
20:46 esskar_ because it is the request
20:46 esskar_ that was just send
20:47 sri even if that was the case it wouldn't make a difference
20:47 esskar_ whyß
20:47 esskar_ ?
20:47 esskar_ i dont see it
20:47 sri because you can change the message before rerun
20:48 esskar_ ?
20:48 esskar_ what message?
20:48 sri so even if you could cache the whole message, you have to rebuild it from scratch
20:48 sri http message of course
20:48 sri it's not even the case though, we build the message in chunks
20:49 sri it could be streaming from a file
20:49 sri a 2gb upload
20:49 sri or more
20:49 esskar_ but you always have transaction
20:49 sri ?
20:49 esskar_ what happens wehen i just say print $c->req
20:49 esskar_ inside the callback
20:49 sri it builds the message
20:49 sri which could go horribly wrong
20:50 sri because it is the second time
20:50 sri print $c->req only works for simple messages in the callback
20:51 esskar_ i think, it's better to support it as much as one can then saying: it's not going to work at all
20:52 sri feel free to do it, but it will require a few hundred tests
20:52 sri http://github.com/kraih/mojo/bl​ob/master/t/mojo/message.t#L662
20:53 sri there is an example for a message that depends on external code
20:53 sri that external code might not be able to run twice
20:54 sri i would love being able to infinitely rerun every transaction, but someone has to write those few hundred tests
20:54 esskar_ i'll give it a try tomorrow
20:55 sri basically every request building test in t/mojo/message.t would have to run 2 or more times
20:55 sri and then you need to catch those that can't
20:57 sri i still don't see how $client->process( $req ) relates to any of this
20:57 sri do you mean my $tx = Mojo::Transaction::HTTP->new(req => $old_tx->req);
20:57 sri ?
20:58 yakudza joined #mojo
21:04 ash_ joined #mojo
21:04 sri esskar_: actually, what you want migth just work already
21:05 sri if you ignore dynamic content
21:05 esskar_ yes, similar
21:05 esskar_ just packed in a lighter api
21:06 esskar_ Mojo::Transaction::HTTP->new(req => $old_tx->req); is "heavy"
21:06 sri umm, what you are doing is very much non standard stuff
21:06 esskar_ $client->process( $req )  or  $client->request( $req ) would be much lighter
21:06 esskar_ is it
21:07 sri i don't think minority use cases deserve a special method
21:07 esskar_ when you get an reponse, not authorized
21:07 esskar_ at is common and standard
21:07 sri not really
21:07 esskar_ that you can just rerun the same request
21:07 sri basic and digest are pretty much dead
21:07 esskar_ with addional headers
21:08 sri this is totally not everyday stuff
21:08 esskar_ perl is also dead ;=
21:08 esskar_ ;-)
21:08 esskar_ lwp supports it
21:09 esskar_ its not straight forword
21:09 esskar_ but possible
21:09 esskar_ i will make it work ;9
21:09 esskar_ (i hope)
21:10 sri lwp can't support the dynamic content cases
21:10 sri or doesn't it allow a callback for content generation?
21:12 esskar_ sure
21:12 esskar_ bnot
21:12 esskar_ not
21:12 esskar_ my idea is, the store the request in a file after it has been send
21:13 sri and then?
21:13 esskar_ rerun would be, read it from a file as it is
21:13 esskar_ and/or change the stuff, that is to changed
21:14 sri you do realize you have to do it for every request?
21:14 esskar_ sure
21:14 sri that means each and every request would be written to disk first
21:15 sri it would be slow as hell
21:15 esskar_ if it has content larger than x (maybe)
21:15 esskar_ then
21:15 sri what about a chunked request?
21:16 esskar_ what about it?
21:16 sri pushing realtime news infinitely
21:16 sri reverse long polling
21:16 esskar_ pff ;-)
21:17 sri btw. the test i linked to above is the perfct example
21:17 sri it can't run twice
21:17 sri just tested, won't buid twice
21:17 sri *build
21:17 sri *perfect
21:18 esskar_ give me tomorrow
21:18 esskar_ no worries so far
21:19 esskar_ i should be possible (some how)
21:19 esskar_ i did it for an reverse proxy i once build in c++
21:19 esskar_ its the same thing actually
21:19 esskar_ to user sends a huge request
21:20 esskar_ the rev proxy gets it
21:20 esskar_ tries to send it to the server
21:20 sri not really
21:20 esskar_ i know
21:20 esskar_ just figured
21:20 esskar_ thats why i stopped )
21:20 esskar_ talking
21:20 esskar_ anyways
21:21 esskar_ i will think about it
21:21 esskar_ good night
21:21 sri browsers can do it because requests are not dynamic
21:21 sri nn
21:24 tempire_ I previously thought Catalyst's method of having routes defined in a corresponding controller was a good way of doing things.  What cooler thing is there than to have Class::MOP analyze all the modules on startup?
21:24 tempire_ but
21:24 tempire_ I repent.
21:24 tempire_ I was blind.
21:24 tempire_ But now I see.
21:28 tempire_ All glory to the hypnotoad.
21:29 sri ALL GLORY TO THE HYPNOTOAD!
21:39 DaTa http://urlchen.de/0Uo0 a bit powering the revolution :)
21:40 sri viva la revolucion!
21:42 yakudza left #mojo
21:43 perlite left #mojo
21:44 perlite joined #mojo
22:02 tempire_ left #mojo
22:09 ltriant joined #mojo
22:25 Htbaa left #mojo
22:38 vti what's happening?
22:49 koban left #mojo
23:02 und3f is now known as und3f[A]
23:04 ana Dexter was awesome, next episode will be the bees knees :}
23:16 ^stereonaut left #mojo
23:17 stereonaut joined #mojo
23:20 spleenjack left #mojo
23:22 fod joined #mojo
23:24 ash_ left #mojo
23:24 ana left #mojo
23:24 ana joined #mojo
23:37 kimoto joined #mojo
23:42 ana left #mojo
23:55 ashleydev left #mojo
23:55 ashleydev joined #mojo

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary